... andsentiment words (Mei et al., 2007; Titov and Mc-Donald, 2008; Zhao et al., 2010; Li et al., 2010b),where a topic is a cluster of words, which is differ-ent from our fine-grained topic -word extraction. Jin ... lexicon extraction, Y ∈ {1, 2, 3}, whereyi= 1 denotes the corresponding word wia sen-timent word, yi= 2 denotes wia topic word, andyi= 3 denotes wineither a sentiment nor topic word. ... forsentiment lexicon extraction. Table 3 shows the comparison results on topic lex-icon extraction. From the table, we can observe thatdifferent from the sentiment lexicon extraction task,the...